Results of a Geophysical Test of Low-Angle Dip on the Seismogenic Dixie Valley Fault, Nevada

The recognition of active low-angle normal faults (< 30 degrees) and the possibility that they may rupture in large earthquakes has direct implications for seismic hazard analysis. Yet, controversy resides in whether low-angle normal faults can initiate and be active at shallow dip (Anderson, 1942). The recent geologic observations of Caskey et al. (1996) along the rupture trace of the 16 December 1954 Ms 6.8 Dixie Valley earthquake (above) indicate a near surface fault dip as low as 25 degrees E along the central portion of the rupture trace. The implication is that the Dixie Valley earthquake may be the first historic example of a large surface rupturing earthquake on a low-angle normal fault. Previous seismological data from the earthquake (Doser, 1986) are unable to confirm or refute this hypothesis, because the Dixie Valley event wave forms are contaminated by those from the preceding Fairview Peak (Ms7.2) earthquake, which occurred 4 minutes earlier. Likewise, previous seismic refraction and gravity studies in southern Dixie Valley have not resolved the subsurface geometry of the Dixie Valley fault.
(Local directory containing DEM data and images; public copy of a C program to convert DEM files to raw 16-bit images.)

This map shows our main geophysical transects highlighted in red. We conducted a multi-element geophysical study to determine the subsurface geometry of the Dixie Valley fault along the 1954 rupture trace in southern Dixie Valley, and hence to test whether the Dixie Valley earthquake provides a true example of a large historical seismic event on a low-angle normal fault. Our most intensive seismic surveys took place between Willow Canyon and Highway 121 on Cattle Road.

Raw data from receivers extending 700 m east from the fault, with a 15 pound explosion in 3 holes 1.45-2.17 km to the east at the highway. A vertically-propagating headwave climbs up the east dip of the basin bottom to arrive simultaneously at all receivers. Thus the fault dip must equal the refraction critical angle. At these depths in the Great Basin, the basin:basement velocity contrast must be at less than 1:2, so the dip of the fault is 30 degrees or less. Further, the straight simultaneous arrival shows the fault surface is not significantly corrugated in the dip direction, for more than 700 m distance east of the rupture, and 350 m depth.
(Local links to a script that made the plot, and a script that converted all the Bison SEG-2 records to binary float.)

This section shows synthetic acoustic waves from a blast at Highway 121, climbing a 30 degree dipping Dixie Valley fault plane at Willow Canyon. Click here to see a 1.9 Mb animated GIF showing acoustic propagation to 1.5 seconds, or here for a 1.9 Mb Quicktime animation. The basin velocity is assumed to be 2 km/s, and the basement is taken to be at 4 km/s. The horizontal headwave-front appears near the surface where the fault daylights, propagating vertically and arriving just after 0.75 s time.
(Local links to a script that generates and plots the synthetic, and an Adobe Photoshop actions archive that converts the PostScript plots to GIF files. Animated GIF prepared from multiple GIF files with Gif Builder; Quicktime conversion by Graphic Converter.)

Comparison of data against acoustic synthetics for the two long-range 15-lb shots on Cattle Rd. 610 m east of Highway 121 (Shot 60, top row), and at Highway 121 (Shot 101, bottom row). The Shot 60 data have been low-pass filtered; and sign reversals of a few data traces have not been corrected. Maximum offset of Shot 60 is about 2775 m. The horizontal first arrival is earlier in the data than in the synthetics because deeper basin velocities are probably faster than the 2 km/s assumed for the synthetics. The 0.25 s time difference between the arrivals from the two shots suggests an average basin velocity east of Highway 121 of about 2.4 km/s. The receiver spread extending only 720 m east of the Dixie Valley fault was not sensitive to any possible higher-angle offset of the basin bottom on the Piedmont fault (1100 m east of the Dixie Val. fault). This possible offset, shown in the model section at left, does not produce any difference between the center and right columns of the synthetics in the figure above.
(Local links to scripts that generate and plot the Shot 101 and Shot 101 Piedmont synthetics, and the Shot 60 and Shot 60 Piedmont synthetics. Shots 101 and 60 data arranged and plotted by this script, Shot 60 data filtered and plotted by this script.)

The medium-resolution seismic reflection data provide constraints on the geometry of the shallowly dipping Dixie Valley fault below and east of the Piedmont fault. Above is a stack of 130 records from 5-lb shots in 3 m holes, employing trace editing, F-K and bandpass filtering, AGC, CDP velocity analysis and stacking, and post-stack Stolt migration. Record processing was done in the Landmark ProMAX system, generously donated to UNR by Wm. Lettis & Assoc.; stacking and migration was done in the RG Seismic Processing System. This animated GIF shows the stack enhancement and migration sequence. The east-dipping Dixie Valley fault and some details of basin stratigraphy, like the Tertiary basalt top, are clear in the upper second of two-way travel time.
(Local directory of SEG-Y files used by ProMAX, processing flow descriptions, and SEG-Y stack results. Local links to the directory of CV stack results, and the directory of stack results; scripts that perform the CV stack and stack; and scripts that perform coherency enhancement, and migrate the stack. There are also the script for dip filtering, and a parameter file for the psmat program that plots the stack.)

While the stack shows the likely rollover anticline against the Dixie Valley fault in the basin sediment and volcanic deposits, it could not image below the high-velocity basalt layer. A prestack Kirchhoff migration of the same F-K filtered records, above, does manage to image the fault below all the but the most reflective regions of the basalt top. The prestack migration uses the Lumley-Claerbout operator antialiasing criteria, and thus represents a longer-wavelength view of the reflectivity in this section. This animated GIF shows the prestack migration enhancement sequence.
(Local links to the directory of prestack results; and scripts that perform prestack migration and coherency enhancement. There are also the script for dip filtering, and a parameter file for the psmat program that plots the migration.)

Combining the post-stack (color) and prestack (gray) migrations show that the prestack image has the deeper and more prominent reflectors detailed by the post-stack processing. A line-drawing interpretation of the combination (below) suggests major constraints on significant fault and basin structure. East-dipping lines trend directly toward the surface outcrop of the Dixie Valley Fault at the upper left corner, with parallel reflections below suggesting the 25-30 degree east-dipping foliation in the granite footwall. The subhorizontal or west-dipping basin sediments terminate at the fault with onlap. In addition, the thick gray lines of the generalized interpretation suggest a rollover anticline, as would be associated with any listric normal fault.

Models of bedrock depth from magnetic (top) and gravity (middle) data from a profile extending from the mouth of East Job Canyon east-northeast to Highway 121, then east on Settlement Road. The cross section of model basin depths (bottom) has no vertical exaggeration. Magnetic modeling was performed in Northwest Geophysical Associates' GM-SYS pakage (kindly donated to UNR by the W. M. Keck Foundation); gravity inversion was with the method of Talwani et al. (1959). Using a large -1.0 g/cc density contrast for the basin fill relative to the basement, the gravity basement is 1 km shallower than the not-well-controlled magnetic basement. Decreasing the contrast to -0.7 or -0.6 g/cc would allow the two interpretations to match better, and both would show about a 30 degree dip for the Dixie Valley fault at the west wall of the basin.
(Public FTP directory holding Excel 5 and tab-delimited versions of all the March 1998 gravity, data-reduction notes, and a script executing the Talwani inversion. Public FTP directory holding notes on errors and modeling, all March 1998 magnetics data as an Excel spreadsheet, and GIF snapshots of GM-SYS models for the Cattle Rd. and Settlement Rd. profiles.)

Model of bedrock depth from gravity data (top) from our main profile extending from the mouth of Willow Canyon east to Highway 121, then east on Settlement Road to the Clan Alpine foothills. The cross section of model basin depths (middle) has no vertical exaggeration. Here we used a regional average density contrast profile for the basin fill relative to the basement, and show the seismic line drawing to scale (bottom).

Migrated brute stack of high-resolution reflection records collected with 100 Hz geophone strings within 130 m of the rupture. Stacking and subsequent Stolt migration used a 670 m/s shallow constant velocity, providing geometric corrections only, with no filtering. Depth conversion with this velocity provides a section with no vertical exaggeration to 67 m depth. The bedrock-alluvium reflector dips at 28 degrees, after adding a 4 degree surface slope to the dip in the section.
(Local links to a script that performs the stack, and a script that migrates and plots the stack.)

Line-drawing interpretation of the migrated high-resolution brute stack (lower), with time-domain elctromagnetic sounding results for the profile (upper). TEM lower-layer resistivity falls for the soundings penetrating into the granite below the Dixie Valley fault. Groundwater trapped in the fracture sets aligned with the fault lowers the resistivity, as well as showing up as east-dipping reflectivity in the high-resolution seismic stack.

References

• Anderson, E. M., 1942, The Dynamics of Faulting, 1st ed.: Oliver and Boyd, Edinburgh, 183 pp.
• Caskey, S. J., S. G. Wesnousky, P. Zhang, and D. B. Slemmons, 1996, Surface faulting of the 1954 Fairview Peak (Ms=7.2) and Dixie Valley (Ms=6.9) earthquakes, central Nevada: Bull. Seism. Soc. Am., v. 86, p. 761-787.
• Doser, D., 1986, Earthquake processes in the Rainbow Mountain-Fairview Peak-Dixie Valley, Nevada region 1954-1959: J. Geophys. Res., v. 91, p. 12572-12586.
• Talwani, M., Worzel, J. L., and Landisman, M. G., 1959, Rapid gravity computations for two-dimensional bodies with application to the Mendocino submarine fracture zone: Journal of Geophysical Research, v. 64, p. 49-59.